Extremely smart reverse polarity protection

[peranders]

Since someone brought this up I just have to tell the world about this very smart solution.

http://www.infineon.com/cmc_upload/migrated_files/document_files/Application_Notes/36348.pdf

See the last page.

If you have cases where you want full reverse polarity protection but can't have the diode Elvis because of lot's of ampere, this solution makes a diode with perfect blocking properties AND only 5-9 mohms in losses in forward connection.

10 A => 50 mW in heat! It's too good to be true.

The BTS650 is capable of 120 A peak in TO220!!

I have tested this circuit myself and it works.

 

[TwoSpoons]

Works with any MOSFET with Vgth < supply. Can be configured with either P or N channel devices, N channel preferred because of lower Rds for same die size.

Used it myself in designs powered by a single cell, where you cannot afford a 0.6V diode drop!

 

[peranders]

Notice though that the BTS650 is an IC but you are quite right any mosfet can be connected backwards sort of speaking.

If only Vgs is positive (for a N-channel) drain currents can flow in either direction but of cource you will only get 0.6 max blocking voltage in the "wrong" direction but you will have losses in milliohms in both directions when the mosfet is turned on.

 

[Upupa Epops]

Very interesting circuit, P-A. But is useable only up to 60 V and is made only for one polarity. Are you designed output protections ? Look at IRFBA90N20D.

 

[peranders]

It is usable up to the voltage for the BTSxxx. It works as a true reverse polarity protection with extremely low losses and up to +- Umax (70 V for BTS660).

 

[jellochaos]

I 'm making a zombie thread but I found this in my endless search for the most efficient AC-DC converter. The application note in the first post interests me. But it seems that the link is old and doesn't work. I tried to find it but I failed. Does anyone of the people that posted here 6 years ago happen to have it saved?